Performing a query with a decimal field on Oracle using the fcSDK can throw an error:

OCI-22053: overflow error

One particular customer saw this error when viewing a ClearContracts contract in fcClient.
The total_net column is defined as decimal (19,4), meaning that there should only be 4 digits after the decimal point. But, the data being stored was 7480.2199999999999999999999878

It looks like the precision of the decimal in Oracle is greater than the precision capable in ADO.NET 1.1.

There is a workaround in ado.net 2.0, but the fcSDK is not yet compiled with .net 2.0. we are working on this - it should ship later in 2007.

One workaround may be to seperatly query for this field using ado directly. This may work.

Otherwise, the workaround may be to fix the data in the database.
The column is defined as decimal (19,4), but clearly, the data is being stored with higher precision than this.

I would suggest looking into how this data is getting there in the first place, and fix that.
Else, "clean up" the data, so its at a proper precision. You could use a trigger or some other batch process to do this.

As an example of how to reproduce:
create a new column of decimal (37, 25) and setting its value to 7480.2199999999999999999999878 using sql plus.
Then query for this contract using the fcSDK, and you will get the overflow error.

• Dovetail SDK – Dovetail Software Development Kit for Clarify